Findings show that sand ramps cannot easily be interpreted in terms of a simple model of fluctuating palaeoenvironmental phases from aeolian dominated to soil/fluvial dominated episodes. They accumulate quickly (perhaps in < 5 ka), probably in a single phase before becoming relict. Based on the evidence from Soldier Mountain, they appear strongly controlled by a 鈥榳indow of opportunity鈥?when sediment supply is plentiful and cease to develop when this sediment supply diminishes and/or the accommodation space is filled up. Contemporary observations of stone movement both on rock and sandy sloping surfaces in the Mojave region indicate movement rates in the order of 0.6 and 11 mm yr鈭?#xA0;1, which is insufficiently fast to explain how stone horizons could have been moved across and been incorporated into sand ramps on multiple occasions. Stone horizons found within the aeolian sediments lack evidence for soil development and are interpreted as very short-term events in which small streams moved and splayed discontinuous stone horizons across the sand ramp surface before aeolian deposition resumed. Surface stone horizons may form by creep from mountain slope sources across sand ramps but require enhanced speed compared to measured rates of runoff creep. We propose the mechanism of fluvio-aeolian creep. Our study suggests that current models of alternating aeolian and colluvial deposition within sand ramps, their palaeoenvironmental significance and indeed how sand ramps are distinguished from other dune forms require amendment.